Microelectronic controller design for pole balancing problem

One reason for the present upsurge of interest in intelligent control is that current generations of control systems are incapable, to a greater or lesser extent, of dealing with problems characterized by a certain complexity. Fortunately, human operators (HO) are often experts in keeping the complex control systems on the right track. Among intelligent controller design techniques, neurocontrol is a dynamic research field that has attached considerable attention from the scientific and control engineering communities in the last few years. In this paper, the template learning neurocontrol approach has been used for controller design where the HO, the most successful intelligent controller available until now, is the template controller. This approach has been applied to a pole balancing problem. Several feedforward neural network structures had been used to implement the controller and their relative advantages and disadvantages were investigated. The resulting controller succeeded in balancing the pole for a reasonable time. The approach seemed to be useful in building complex nonlinear systems controllers where there was no need for system models and ease of development.